Image de couverture de livre

Speculative Decoding Scaling Laws (SDSL): Throughput Optimization Made Simple

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Bozorgkhoo, Amirhossein, Molybog, Igor
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language Information Theory Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Block Verification Accelerates Speculative Decoding
par: Sun, Ziteng, et autres
Publié: (2024)

SpecTr: Fast Speculative Decoding via Optimal Transport
par: Sun, Ziteng, et autres
Publié: (2023)

SODBench: A Large Language Model Approach to Documenting Spreadsheet Operations
par: Indika, Amila, et autres
Publié: (2025)

BCAmirs at SemEval-2024 Task 4: Beyond Words: A Multimodal and Multilingual Exploration of Persuasion in Memes
par: Abaskohi, Amirhossein, et autres
Publié: (2024)

REST: Retrieval-Based Speculative Decoding
par: He, Zhenyu, et autres
Publié: (2023)

Multi-Draft Speculative Sampling: Canonical Decomposition and Theoretical Limits
par: Khisti, Ashish, et autres
Publié: (2024)

Scaling Speculative Decoding with Lookahead Reasoning
par: Fu, Yichao, et autres
Publié: (2025)

Decoding Speculative Decoding
par: Yan, Minghao, et autres
Publié: (2024)

A Simple Model of Inference Scaling Laws
par: Levi, Noam
Publié: (2024)

Speeding up Speculative Decoding via Sequential Approximate Verification
par: Zhong, Meiyu, et autres
Publié: (2025)

Semantic Faithfulness and Entropy Production Measures to Tame Your LLM Demons and Manage Hallucinations
par: Halperin, Igor
Publié: (2025)

Greedy Multi-Path Block Verification for Faster Decoding in Speculative Sampling
par: Thomas, Rahul, et autres
Publié: (2026)

SPEX: Scaling Feature Interaction Explanations for LLMs
par: Kang, Justin Singh, et autres
Publié: (2025)

LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding
par: Samarin, Alexander, et autres
Publié: (2026)

Accelerating Retrieval-Augmented Language Model Serving with Speculation
par: Zhang, Zhihao, et autres
Publié: (2024)

Speculative Decoding Across Languages
par: Paudel, Nirajan, et autres
Publié: (2026)

L$^2$M: Mutual Information Scaling Law for Long-Context Language Modeling
par: Chen, Zhuo, et autres
Publié: (2025)

Future Validity is the Missing Statistic: From Impossibility to $Φ$-Estimation for Grammar-Faithful Speculative Decoding
par: Nie, Wenhua, et autres
Publié: (2026)

Proposal and study of statistical features for string similarity computation and classification
par: Rodrigues, E. O., et autres
Publié: (2026)

Efficient Learned Data Compression via Dual-Stream Feature Decoupling
par: Ma, Huidong, et autres
Publié: (2026)

Effective Context in Transformers: An Analysis of Fragmentation and Tokenization
par: Fesharaki, Amirmehdi Jafari, et autres
Publié: (2026)

RateQuant: Optimal Mixed-Precision KV Cache Quantization via Rate-Distortion Theory
par: Zuo, Fei, et autres
Publié: (2026)

Filtering Beats Fine Tuning: A Bayesian Kalman View of In Context Learning in LLMs
par: Kiruluta, Andrew
Publié: (2026)

A Rate-Distortion Framework for Summarization
par: Arda, Enes, et autres
Publié: (2025)

Attention with Markov: A Framework for Principled Analysis of Transformers via Markov Chains
par: Makkuva, Ashok Vardhan, et autres
Publié: (2024)

Theoretical guarantees on the best-of-n alignment policy
par: Beirami, Ahmad, et autres
Publié: (2024)

Multi-turn Training with Basic Human Feedback Helps Little on LLM Reasoning
par: Liu, Qiang, et autres
Publié: (2025)

Understanding Factual Recall in Transformers via Associative Memories
par: Nichani, Eshaan, et autres
Publié: (2024)

An Information-theoretic Multi-task Representation Learning Framework for Natural Language Understanding
par: Hu, Dou, et autres
Publié: (2025)

InfAlign: Inference-aware language model alignment
par: Balashankar, Ananth, et autres
Publié: (2024)

A Mathematical Theory for Learning Semantic Languages by Abstract Learners
par: Liao, Kuo-Yu, et autres
Publié: (2024)

Transformers on Markov Data: Constant Depth Suffices
par: Rajaraman, Nived, et autres
Publié: (2024)

What Makes the Preferred Thinking Direction for LLMs in Multiple-choice Questions?
par: Zhang, Yizhe, et autres
Publié: (2025)

An Enhanced Text Compression Approach Using Transformer-based Language Models
par: Rahman, Chowdhury Mofizur, et autres
Publié: (2024)

Fundamental Limits of Prompt Compression: A Rate-Distortion Framework for Black-Box Language Models
par: Nagle, Alliot, et autres
Publié: (2024)

MultiTok: Variable-Length Tokenization for Efficient LLMs Adapted from LZW Compression
par: Elias, Noel, et autres
Publié: (2024)

Cost-aware LLM-based Online Dataset Annotation
par: Elumar, Eray Can, et autres
Publié: (2025)

Iterative Counterfactual Data Augmentation
par: Plyler, Mitchell, et autres
Publié: (2025)

Latent Space Alignment for Semantic Channel Equalization
par: Hüttebräucker, Tomás, et autres
Publié: (2024)

Clinical Reading Comprehension with Encoder-Decoder Models Enhanced by Direct Preference Optimization
par: Nahian, Md Sultan Al, et autres
Publié: (2024)